- Email: [email protected]
A preliminary exploration of plain-film radiography in scapular dyskinesis evaluation
ARTICLE IN PRESS J Shoulder Elbow Surg (2018) ■■, ■■–■■
www.elsevier.com/locate/ymse
ORIGINAL ARTICLE
A preliminary exploration of plain-film radiography in scapular dyskinesis evaluation Kang Chen, MD1, Simin Deng, MD1, Yanhong Ma, MD*, Yelin Yao, MD, Juan Chen, MD, Yuqian Zhang, MD Department of Rehabiliation Medicine, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, Shanghai, China Background: Evaluation of scapular dyskinesis is of clinical interest because it is believed to be associated with pathologies of the shoulder. This study investigated the feasibility of plain-film radiography in evaluating scapular dyskinesis. Methods: Subjects with unilateral disorders of the shoulder (n = 186) who underwent plain-film radiography of bilateral scapulae were divided into 4 categories of scapular dyskinesis patterns according to the Kibler classification and analyzed. Coracoid upward shift distance (CUSD), length of the scapular spine line (LSS), and scapular upward rotation angle (SURA) were measured on the radiographs. Intrarater and inter-rater reliability were tested, and the characteristics of these parameters in each type were analyzed. The differences (d) between bilateral scapulae (d-CUSD, d-LSS, and d-SURA) among the 4 categories were compared. Results: Intrarater and inter-rater reliability were excellent for all parameters. Significant differences between the scapulae were observed in CUSD in type I and in LSS in type II categories. No significant difference in any of the parameters was found in type III. Compared with the other categories, d-CUSD in type I and d-LSS in type II were significantly larger. The cutoff values of d-CUSD and d-LSS were 1.1 mm and 1.2 mm, respectively. No significant difference in d-SURA was found among the 4 categories. Conclusions: The measurement of CUSD, LSS, and SURA on plain-film radiography had excellent reliability. d-CUSD and d-LSS were characteristic parameters of type I and type II, respectively; however, type III had no distinguishing characteristics among the parameters. Level of evidence: Level III; Diagnostic Study © 2018 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. Keywords: Scapular dyskinesis; plain-film radiography; coracoid upward shift distance; length of the scapular spine line; scapular upward rotation angle; Kibler classification
Abnormalities in the positions of the scapulae at rest and during movements have been collectively termed scapular dyskinesis,15,20,27 which is associated with many shoulder disorders.8,11,17-19,25 Management of scapular dyskinesis is one The Shanghai Jiao Tong University Affiliated Sixth People’s Hospital Ethics Committee approved the study: Approval No. 2015-43-(1). 1 These authors contributed equally to this work. *Reprint requests: Yanhong Ma, MD, Department of Rehabilitation Medicine, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, 600 Yishan Rd, Shanghai, 200233, China. E-mail address: [email protected] (Y. Ma).
of the strategies of physiotherapy in disorders of the shoulder; therefore, appropriate evaluation is important in the management of scapular dyskinesis. Kibler et al20 introduced a classification system involving 4 categories for scapular dyskinesis according to the abnormal movements of the scapula in 3 planes: Type I: abnormal scapular movements in the sagittal plane with anterior and posterior tilts Type II: abnormal scapular movements in the transverse plane with internal and external rotations
1058-2746/$ - see front matter © 2018 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. https://doi.org/10.1016/j.jse.2017.12.032
ARTICLE IN PRESS 2
K. Chen et al. Type III: elevation of the scapula and abnormal movements in the coronal plane with upward and downward rotations Type IV: symmetrical bilateral scapula
This classification system is widely used in clinical settings15,16,20-22 but is based on visual observations.20 In many cases, the soft tissues overlying the scapulae, such as adipose tissue and muscles, may make the identification difficult,30 and this method cannot quantify the severity of scapular dyskinesis. Laboratory instruments, such as inclinometers and electromagnetic or optoelectronic tracking systems, can quantify scapular kinematics in the 3 dimensions.3,8,10,24,28 These methods allow for accurate measurements; however, the technical difficulties and requirements of the equipment render them unavailable in the clinical settings. Three-dimensional (3D) wing computed tomography (CT) and magnetic resonance imaging (MRI) have been explored in evaluating scapular dyskinesis.21-23 However, CT and MRI are both performed in the decubitus position, which relieves the scapula of gravity and the resulting movements. Whether scapular kinematics in decubitus are the same as those in the erect position is unclear. Endo et al 6,7 introduced 3 parameters on plain-film radiography to represent 3 kinds of scapular rotational tilts. The length of the scapular spine line (LSS) represents internal and external rotations, coracoid upward shift distance (CUSD) represents anterior and posterior tilts, and the scapular upward rotation angle (SURA) represents upward and downward rotations. Plain film is advantageous because of lower radiation, fewer technical requirements, and inexpensive and easily available equipment. We hypothesized that LSS, CUSD, and SURA can be used to evaluate scapular dyskinesis. Therefore, the purpose of this study was to investigate the characteristics of radiographic parameters in the 4 patterns of scapular dyskinesis and the feasibility of plain-film radiography in evaluating scapular dyskinesis.
Materials and methods Subjects and demographics This was a prospective, diagnostic case series involving repeated measurements. All subjects received information regarding the objectives of the study and provided their written consent for participation. Subjects who had unilateral shoulder disorders were recruited from the Department of Rehabilitation Medicine by 1 investigator (K.C.). Exclusion criteria were (1) a history of fracture or deformity of the scapula, clavicle, humerus, or ribs; (2) scoliosis or kyphosis with visible rib hump in neutral erect posture; (3) unequal lengths of lower extremities; (4) stature >1.8 m or body mass index >26 kg/m2;12 (5) subjects whose scapular pattern could not be determined by 3
Table I
Demographic data of subjects No. or mean ± SD (range)
Variable
(N = 186) Sex Women Men Age, yr Mass, kg Height, m BMI, kg/m2 Affected side Left Right Dominant side Left Right Dominant side Affected Unaffected Duration, weeks
115 71 45.74 ± 10.20 (22-68) 59.60 ± 6.94 (46-80) 1.64 ± 0.07 (1.50-1.80) 22.14 ± 1.93 (17.47-25.96) 73 113 15 171 110 76 16.82 ± 13.22 (1-48)
SD, standard deviation; BMI, body mass index.
Table II
Main disorders in the affected shoulders
Main disorder of affected side
No. (N = 186)
Subacromial impingement syndrome Partial-thickness rotator cuff tear Adhesive capsulitis of shoulder Long head of biceps tendon tendinitis Shoulder osteoarthritis Rotator cuff tendonitis Superior labrum anterior posterior tear
59 56 23 15 13 10 10
raters; and (6) obvious variations or bony spurs of the coracoid and acromion as seen on x-ray images. We enrolled 205 subjects who underwent radiographic examination and clinical classification of scapular dyskinesis. After the radiographic examination, 19 individuals were excluded for the following reasons: (1) obvious acromion spurs in 8 and morphologically variable coracoid processes in 2; (2) mild thoracic scoliosis in 3; and (3) the quality of imaging was inadequate in 6. Finally, 186 individuals were included, and their demographic data are summarized in Tables I and II.
Clinical evaluation for scapular dyskinesis Scapular patterns were identified based on visual palpation because it has the advantages of identification of bony landmarks and reliability.15 To expose both the scapulae, men bared to the waist and women wore halter tops. The raters stood behind the subject and documented the initial rating by visual inspection. Then they placed their hands on the subject’s scapulae for further identification
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation
3 posterior end of the acromion of the acromioclavicular joint.6,7 LSS was defined as the length of the scapular spine line (Fig. 2, A). CUSD was defined as the distance between the scapular spine line and the upper border of the coracoid process (Fig. 2, A). If the upper border of the coracoid process was above the scapular spine line, the value was denoted with a positive (+) sign and in other cases, a minus (–) sign was attached. The angle between the scapular spine line and the horizontal was defined as SURA (Fig. 2, B). If the LSS was above the horizontal, the value was denoted with a positive (+) sign and in other cases, it was denoted with a minus (–) sign.6,7
Reliability Figure 1 Schematic diagram of the visual-based palpation process in evaluating scapular dyskinesis.
(Fig. 1).15,16 The scapular patterns were classified according to the previously described Kibler classification.15,16,20,27 Three raters with previous experience with assessments of scapular dyskinesis4 assessed the subjects together and were allowed discussions to derive a unanimous decision. Before the study began, the raters were trained for 2 months using visual and verbal presentations, and they practiced the assessments in 10 subjects in a preliminary study. They had the same criteria for evaluation and were allowed discussions; therefore, they were able to provide agreements on all assessments. If a scapular pattern could not be determined because of a disagreement among the raters, the subject was excluded.
A priori power analysis showed that 22 was the minimum required number of participants to establish a reliability coefficient of 0.80 with α = 0.05 and 1 – β = 0.80.29 A computer-generated random number table was used to randomly select 22 radiographs from the 186 plain films for assessing the reproducibility of measurements. These measurements were observed by 2 measurers, with an interval of 1 week between the first and second measurements. The radiographic measurers were blinded to the results of the clinical evaluation.
Data processing The final value of a parameter was the average of 3 measurements. d-CUSD, d-LSS, and d-SURA were the differences in the values between the unaffected and affected sides.
Radiographic technique
Statistical analysis
Immediately after the clinical evaluation, subjects underwent anteroposterior (AP) radiography of both scapulae. The subject stood before the imaging plate, facing the x-ray tube, in the same erect position as that in the clinical evaluation. The sternal angle was selected as the radiographic center. The distance between the film and the x-ray tube was set at 1.5 m. The tube was directed perpendicular to the frontal axis of the thorax. All radiographies were obtained by 1 experienced image technologist.
SPSS 19.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA) software were used for statistical analyses. The normality of the quantitative data was tested using the Kolmogorov-Smirnov test. Intraclass correlation coefficient (ICC) was used to test intrarater and inter-rater reliability, which was denoted as poor (0.00-0.20), fair (0.21-0.40), good (0.41-0.75), or excellent (>0.75) reliability.9 The 95% confidence intervals (CIs) of the ICC of each parameter of the unaffected side and the affected side were computed. The standard error of measurement (SEM) and the minimal detectable change (MDC) for 95% CIs were calculated for each parameter using the following formula:
Radiographic parameters measurements Radiographic plain film was saved in DICOM (Digital Imaging and Communications in Medicine; National Electrical Manufacturers Association, Rosslyn, VA, USA) format in a picture archiving and communication system. Digimizer 4.3.5 software (MedCalc Software bvba, Ostend, Belgium) was used to measure the distances and angles.13 During the measurement, the radiographic digital images were magnified to clearly visualize the inner and outer points (Fig. 2). The right scapula was measured first, followed by the left. Each measurement was repeated 3 times. The measurer first marked a referential line, named the scapula spine line, on the image. The inner point of this line was the intersection of the upper border of the scapular spine and the floor of the supraspinatus fossa, and the outer point was the
SEM = SDpooled × 1 − ICC , MDC = SEM × 2 × 1.96, where SDpooled was the pooled standard deviation.14 The side-to-side differences in the radiographic parameters were tested by the paired Student test. One-way analysis of variance was used to compare d-CUSD, d-LSS, and d-SURA between the 4 types of scapular dyskinesis, and the least significant difference test was performed as a post hoc test when needed. The parameters with significant differences in d-CUSD and d-LSS on analysis of variance were further analyzed using the receiver operating characteristic (ROC) curves. Statistical significance was defined as a P value of < .05.
ARTICLE IN PRESS 4
K. Chen et al.
Figure 2 Schematic diagram of measuring the radiographic parameters. Anteroposterior radiograph of a shoulder at rest is shown. The red dots represent the inner and outer points of the scapular spine line. (A) Length of the scapular spine line (LSS) was defined as the length of the scapular spine line. Coracoid upward shift distance (CUSD) was defined as the distance between the scapular spine line and the upper border of the coracoid process. (B) Scapular upward rotation angle (SURA) was defined as the angle between the scapular spine line and the horizontal.
Results
IV patterns (Fig. 3). Representative radiographs are shown in Fig. 4.
Clinical evaluation In the 186 subjects, the numbers of subjects with scapular dyskinesis type I, II, III, and IV were 28, 41, 71, and 46, respectively.
Reliability of the radiographic measurement The reliability, SEM, and MDC of parameters are summarized in Tables III and IV. Intrarater and inter-rater reliabilities for all parameters in the unaffected and affected sides were excellent.
Side-to-side comparisons of the radiographic parameters
Parameter CUSD, mm LSS, mm SURA, °
Significant differences in d-CUSD and d-LSS were observed among the 4 types (P < .001). The post hoc test showed that d-CUSD in type I was significantly larger compared with the other types (type II: P = .001; type III: P < .001; type IV: P < .001; Fig. 5, A). d-LSS in type II was significantly larger than that in the other types (P < .001) (Fig. 5, B). There was no obvious difference in d-SURA among the 4 types (P = .254; Fig. 5, C).
Predictive values of d-CUSD for type I and d-LSS for type II
CUSD in type I and LSS in type II indicated a significant difference between the unaffected and affected sides (CUSD: 2.8 ± 2.5 mm; LSS: 3.9 ± 4.0 mm). None of the parameters showed significant side-to-side differences in type III and type Table III
Comparison of d-CUSD, d-LSS, and d-SURA among the 4 types
In view of the significant correlations shown by d-CUSD and d-LSS with type I and type II patterns, respectively, ROC curves were drawn to evaluate the predictive ability of these
Intrarater reliability of radiographic parameters in the affected and unaffected sides Side Affected Unaffected Affected Unaffected Affected Unaffected
Measurement 1
Measurement 2
Mean ± SD
Mean ± SD
0.5 ± 3.9 1.0 ± 4.4 70.7 ± 5.6 69.4 ± 6.4 7.5 ± 6.2 9.4 ± 7.6
0.5 ± 3.9 1.0 ± 4.3 70.1 ± 5.2 69.1 ± 6.2 8.4 ± 7.0 9.7 ± 7.5
ICC (95% CI)
SEM
MDC95
0.999 (0.998-1) 1 (0.999-1) 0.982 (0.957-0.993) 0.992 (0.980-0.997) 0.962 (0.909-0.984) 0.997 (0.992-0.999)
0.1 0 0.7 0.6 1.3 0.4
0.4 0 2.0 1.6 3.5 1.1
ICC, intraclass correlation coefficient; CI, confidence interval; SEM, standard error of measurement; MDC95, minimal detectable change for 95% confidence interval; SD, standard deviation; CUSD, coracoid upward shift distance; LSS, length of the scapular spine line; SURA, scapular upward rotation angle.
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation Table IV
5
Inter-rater reliability of radiographic parameters in the affected and unaffected sides
Parameter CUSD, mm LSS, mm SURA, °
Side Affected Unaffected Affected Unaffected Affected Unaffected
Measurer 1
Measurer 2
Mean ± SD
Mean ± SD
0.5 ± 3.9 1.0 ± 4.4 70.7 ± 5.6 69.4 ± 6.4 7.5 ± 6.2 9.4 ± 7.6
0.4 ± 4.2 1.0 ± 4.4 71.1 ± 5.8 69.9 ± 6.2 8.2 ± 7.1 9.8 ± 7.4
ICC (95%CI)
SEM
MDC95
0.992 (0.981-0.997) 1 (0.999-1) 0.982 (0.956-0.992) 0.989 (0.974-0.996) 0.958 (0.899-0.983) 0.996 (0.991-0.998)
0.4 0 0.8 0.7 1.4 0.5
1 0 2.1 1.8 3.8 1.3
ICC, intraclass correlation coefficient; CI, confidence interval; SEM, standard error of the measurement; MDC95, minimal detectable change for 95% confidence interval; SD, standard deviation; CUSD, coracoid upward shift distance; LSS, length of the scapular spine line; SURA, scapular upward rotation angle.
2 parameters. The cutoff values of ROC curves and area under curve (AUC) are reported in Table V. The parameter with values smaller than the cutoff value was considered to be negative, and the parameter with values larger than the cutoff was considered positive and diagnostically significant.
Discussion With the deep understanding of the relationship between scapular kinematics and shoulder disorders, it is imperative to study the treatment modality of scapular dyskinesis.10,18,19 Although assessment of symptoms or function is often used to evaluate the effectiveness of the treatment modality used in scapular dyskinesis,2,26 few quantitative methods are suitable for clinical evaluation of the changes in scapular dyskinesis. This study shows that plain-film radiography can be used to evaluate scapular dyskinesis. Plain-film radiography is a simple and feasible investigation. Although it is 2D imaging, CUSD, LSS, and SURA can be used to evaluate 3D rotational tilts of the scapula.6,7 In the scapula-lateral view, the distance between the upper end of the coracoid process and the lateral point of the scapular spine line can be considered to be constant.1 Photoshop software (Adobe Systems Inc., San Jose, CA , USA) can be used to rotate the digital image to make the scapular line horizontal.6 CUSD is equal to the product of this constant and the sine value of the anterior/posterior tilt angle in the sagittal plane, and the corresponding change in CUSD would depend on the anterior/posterior tilt angle (Fig. 6, A).7 From a superioraxial view, the length of the scapular spine can be regarded as a constant, LSS is equal to the product of this constant and the cosine value of the internal/external rotation angle, and the corresponding change of the LSS would depend on the internal/external rotation angle (Fig. 6, B).7 The upward or downward rotation of the scapula is in the coronal plane; therefore, the corresponding change in SURA would depend on the angle of the rotation.6,7 As described by Kibler, the type I pattern would be associated with abnormal anterior tilting of the scapula, and the type II pattern would be associated with abnormal scapular
internal rotation.20,27 Some previous studies that observed the scapular pattern using 3D wing CT have proved that subjects with type I exhibit increased anterior tilt and subjects with type II exhibit increased internal rotation.21,22 In this study, CUSD in type I and LSS in type II patterns were significantly smaller in the affected side compared with the unaffected side. Further testing of MDC for the 95% CI indicated that CUSD and LSS could significantly assess anterior tilt and internal rotation, respectively. Similar to the study by Endo et al,6 excellent ICC was achieved for the intrarater and inter-rater reliabilities for all radiographic parameters bilaterally in this study. This may be attributed to the experience of the measurer and standardization of the procedures. Satisfactory reliability is essential for clinical application of such inferences. The AUC reflects the accuracy of a measurement. An AUC of <0.5 is considered random association, whereas an AUC between 0.7 and 0.8 or between 0.8 and 0.9 is considered adequate or excellent.31 Our results have demonstrated that d-CUSD for type I and d-LSS for type II were reasonably accurate (AUC values were 0.783 and 0.752, respectively). Therefore, d-CUSD and d-LSS can be used as the characteristic parameters in the diagnosis of type I and type II patterns. In addition, our study found that when d-CUSD was 1.1 mm, its sensitivity was 0.821 and specificity was 0.69 for identifying type I and nontype I patterns. When d-LSS was 1.2 mm, its sensitivity was 0.756 and specificity was 0.731 for identifying type II and nontype II patterns. The results indicated that d-CUSD ≥1.1 mm and d-LSS ≥1.2 mm have diagnostic significance. Although the Kibler classification system has been accepted as the standard in the evaluation of scapular dyskinesis,15,20,21 some controversies still exist, especially regarding type III. Some investigators believe that type III is associated with excessive upward rotation of the scapula,5,15 whereas other investigators believe that the abnormal movement in type III occurs in multiple planes, including elevation, anterior displacement, or upward rotation.20-22,27 In our study, SURA, which represented upward rotation of the scapula, showed no significant change in subjects with
ARTICLE IN PRESS 6
K. Chen et al.
Figure 3 Side-to-side comparisons of the radiographic parameters for each type of scapular dyskinesis. The range bars show the standard error of measurement. CUSD, coracoid upward shift distance; LSS, length of the scapular spine line; SURA, scapular upward rotation angle.
type III pattern. Similarly, CUSD and LSS showed no significance in type III pattern. We therefore believe that the type III pattern is not simply presented as upward rotation, anterior tilt, or internal rotation of the scapula. The movement
pattern in type III may involve motions other than rotation. It may be necessary to develop other parameters or methods to describe the movements of the scapula in type III pattern or define subtypes in terms of the various abnormal planes
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation
7
Figure 4 Representative radiographs of each type of scapular dyskinesis. (A) Type I: Coracoid upward shift distance (CUSD) in the affected and unaffected sides is –6.1 mm and 2.9 mm, respectively. (B) Type II: Length of the scapular spine line (LSS) in the affected and unaffected sides is 54.7 and 63.6 mm, respectively. (C) Type III: Scapular upward rotation angle (SURA) in the affected and unaffected sides is 10.8 and 4.8°, respectively. (D) Type IV: CUSD and LSS in the affected and unaffected sides are –0.4 mm and –0.8 mm, and 61.6 mm and 61.1 mm, respectively. (E) Type IV: SURA in the affected and unaffected sides is 9.9° and 9.3°, respectively. The black ellipse indicates the affected side.
Figure 5 Comparison of difference (d) for coracoid upward shift distance (d-CUSD), length of the scapular spine line (d-LSS), and scapular upward rotation angle (d-SURA) among the 4 types of scapular pattern are shown. The range bars show the standard error of measurement.
of movements. According to the current findings, a scapular pattern that does not meet the diagnostic criteria of type I or type II may be classified as type III. Scapular dyskinesis can occur when stationary and in motion, only in motion, or only in a static state.4,20 The movements of the glenohumeral joint occur in various directions and are often complex2,19; therefore, the evaluation of scapular
abnormalities during motion is difficult. The current study only explored the evaluation for the scapular dyskinesis in a static position. Compared with dynamic abnormalities, static abnormalities mainly involve the tension and length of soft tissue, and dynamic abnormalities may also involve the abnormalities of the strength of the related muscle.4 One of the limitations of this method is that it is only suitable for evaluating scapular
ARTICLE IN PRESS 8
K. Chen et al. Table V Diagnostic accuracy of difference in coracoid upward shift distance and difference in length of the scapular spine line for specific types of scapular dyskinesis Parameter
Group vs. group
Cutoff value by ROC curve, mm
Sensitivity
Specificity
AUC (95% CI)
P
d-CUSD d-LSS
Type I vs. type II, III, IV Type II vs. type I, III, IV
1.1 1.2
0.821 0.756
0.69 0.731
0.783 (0.717-0.84) 0.752 (0.684-0.813)
<.001 <.001
ROC, receiver operating characteristic; AUC, area under the curve; CI, confidence interval; d-CUSD, difference in coracoid upward shift distance; d-LSS, difference in length of the scapular spine line.
Figure 6 Theoretical analysis of the radiographic parameters. (A) Schema of the coracoid upward shift distance (CUSD) is shown in the scapula-lateral view. The distance between the upper end of the coracoid process and the lateral point of the scapular spine line is labeled l. The line l and the horizontal plane formed the α angle (anterior/posterior angle). Because l was constant, sin α becomes the only variable that determines CUSD. (B) Schema of the length of the scapular spine line (LSS) in the superior-axial view. The length of the scapular spine was labeled n. The angle β (internal/external angle) was formed between the line n and the coronal plane. Because n could be regarded as a constant, cosine β becomes the only variable that determines LSS.
dyskinesis in static states, and patients with abnormalities during movements were not included. There were other limitations in the present study. First, the results of this study were based on a relative small sample of each pattern of scapular dyskinesis. A larger sample size is needed to confirm the universality of the findings in future studies. Second, this study investigated a side-to-side comparison, and the results cannot be applied to evaluate subjects who have bilateral shoulder disorders. Third, the stature and body mass index of subjects were limited because of the limited size of the imaging plate. Fourth, because a clear definition of the line of reference was essential for the measurements, subjects with obvious variations or bony spurs of the coracoid and acromion were excluded; therefore, populations suitable for present radiologic examination are limited.
Conclusion Type I and type II scapular dyskinesis can be evaluated with d-CUSD and d-LSS on plain-film radiography with excellent reliability. However, further explorations are warranted to understand the characteristics of type III.
Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References 1. Alobaidy MA, Soames RW. Evaluation of the coracoid and coracoacromial arch geometry on Thiel-embalmed cadavers using the three-dimensional MicroScribe digitizer. J Shoulder Elbow Surg 2016;25:136-41. http://dx.doi.org/10.1016/j.jse.2015.08.036 2. Carbone S, Postacchini R, Gumina S. Scapular dyskinesis and SICK syndrome in patients with a chronic type III acromioclavicular dislocation. Results of rehabilitation. Knee Surg Sports Traumatol Arthrosc 2015;23:1473-80. http://dx.doi.org/10.1007/s00167-014-2844-5 3. Chu Y, Akins J, Lovalekar M, Tashman S, Lephart S, Sell T. Validation of a video-based motion analysis technique in 3-D dynamic scapular kinematic measurements. J Biomech 2012;45:2462-6. http://dx.doi.org/ 10.1016/j.jbiomech.2012.06.025 4. Deng S, Chen K, Ma Y, Chen J, Huang M. The influence of test positions on clinical assessment for scapular dyskinesis. PM R 2017;9:761-6. http://dx.doi.org/10.1016/j.pmrj.2016.11.011 5. Ellenbecker TS, Kibler WB, Bailie DS, Caplinger R, Davies GJ, Riemann BL. Reliability of scapular classification in examination of professional
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation
6.
7.
8.
9. 10.
11.
12. 13.
14. 15.
16.
17. 18.
19.
baseball players. Clin Orthop Relat Res 2012;470:1540-4. http:// dx.doi.org/10.1007/s11999-011-2216-0 Endo K, Ikata T, Katoh S, Takeda Y. Radiographic assessment of scapular rotational tilt in chronic shoulder impingement syndrome. J Orthop Sci 2001;6:3-10. Endo K, Yukata K, Yasui N. Influence of age on scapulo-thoracic orientation. Clin Biomech (Bristol, Avon) 2004;19:1009-13. http:// dx.doi.org/10.1016/j.clinbiomech.2004.07.011 Fayad F, Roby-Brami A, Yazbeck C, Hanneton S, Lefevre-Colau MM, Gautheron V, et al. Three-dimensional scapular kinematics and scapulohumeral rhythm in patients with glenohumeral osteoarthritis or frozen shoulder. J Biomech 2008;41:326-32. http://dx.doi.org/10.1016/ j.jbiomech.2007.09.004 Fleiss J. The design and analysis of clinical experiments. New York: John Wiley & Sons; 1986. Forthomme B, Crielaard JM, Croisier JL. Scapular positioning in athlete’s shoulder : particularities, clinical measurements and implications. Sports Med 2008;38:369-86. Groh GI, Mighell MA, Basamania CJ, Kibler WB. All things clavicle: from acromioclavicular to sternoclavicular and all points in between. Instr Course Lect 2016;65:181-96. Gupta M, Dashottar A, Borstad JD. Scapula kinematics differ by body mass index. J Appl Biomech 2013;29:6. Habibi E, Soury S, Zadeh AH. Precise evaluation of anthropometric 2D software processing of hand in comparison with direct method. J Med Signals Sens 2013;3:256-61. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 2000;30:1-15. Huang TS, Huang HY, Wang TG, Tsai YS, Lin JJ. Comprehensive classification test of scapular dyskinesis: a reliability study. Man Ther 2015;20:427-32. http://dx.doi.org/10.1016/j.math.2014.10.017 Huang TS, Ou HL, Huang CY, Lin JJ. Specific kinematics and associated muscle activation in individuals with scapular dyskinesis. J Shoulder Elbow Surg 2015;24:1227-34. http://dx.doi.org/10.1016/ j.jse.2014.12.022 Kibler WB. The scapula in rotator cuff disease. Med Sport Sci 2012;57:27-40. http://dx.doi.org/10.1159/000328877 Kibler WB, Sciascia A. The role of the scapula in preventing and treating shoulder instability. Knee Surg Sports Traumatol Arthrosc 2016;24:390-7. http://dx.doi.org/10.1007/s00167-015-3736-z Kibler WB, Sciascia A, Wilkes T. Scapular dyskinesis and its relation to shoulder injury. J Am Acad Orthop Surg 2012;20:364-72. http:// dx.doi.org/10.5435/JAAOS-20-06-364
9 20. Kibler WB, Uhl TL, Maddux JW, Brooks PV, Zeller B, McMullen J. Qualitative clinical evaluation of scapular dysfunction: a reliability study. J Shoulder Elbow Surg 2002;11:550-6. http://dx.doi.org/10.1067/ mse.2002.126766 21. Park JY, Hwang JT, Kim KM, Makkar D, Moon SG, Han KJ. How to assess scapular dyskinesis precisely: 3-dimensional wing computer tomography–a new diagnostic modality. J Shoulder Elbow Surg 2013;22:1084-91. http://dx.doi.org/10.1016/j.jse.2012.10.046 22. Park JY, Hwang JT, Oh KS, Kim SJ, Kim NR, Cha MJ. Revisit to scapular dyskinesis: three-dimensional wing computed tomography in prone position. J Shoulder Elbow Surg 2014;23:821-8. http://dx.doi.org/ 10.1016/j.jse.2013.08.016 23. Sahara W, Sugamoto K, Murai M, Yoshikawa H. Three-dimensional clavicular and acromioclavicular rotations during arm abduction using vertically open MRI. J Orthop Res 2007;25:1243-9. http://dx.doi.org/ 10.1002/jor.20407 24. Shakeri H, Keshavarz R, Arab AM, Ghosheh FT, TalimKhani A. Scapular position and orientation during abduction, flexion and scapular plane elevation phase. Iran Rehabil J 2014;12:22-30. 25. Timmons MK, Thigpen CA, Seitz AL, Karduna AR, Arnold BL, Michener LA. Scapular kinematics and subacromial-impingement syndrome: a meta-analysis. J Sport Rehabil 2012;21:354-70. 26. Turgut E, Duzgun I, Baltaci G. Effects of scapular stabilization exercise training on scapular kinematics, disability, and pain in subacromial impingement: a randomized controlled trial. Arch Phys Med Rehabil 2017;98:1915-23, e1913. http://dx.doi.org/10.1016/j.apmr.2017.05.023 27. Uhl TL, Kibler WB, Gecewich B, Tripp BL. Evaluation of clinical assessment methods for scapular dyskinesis. Arthroscopy 2009;25:12408. http://dx.doi.org/10.1016/j.arthro.2009.06.007 28. van den Noort JC, Wiertsema SH, Hekman KM, Schonhuth CP, Dekker J, Harlaar J. Reliability and precision of 3D wireless measurement of scapular kinematics. Med Biol Eng Comput 2014;52:921-31. http:// dx.doi.org/10.1007/s11517-014-1186-2 29. Walter SD, Eliasziw M, Donner A. Sample size and optimal designs for reliability studies. Stat Med 1998;17:101-10. 30. Wassinger CA, Williams DA, Milosavljevic S, Hegedus EJ. Clinical reliability and diagnostic accuracy of visual scapulohumeral movement evaluation in detecting patients with shoulder impairment. Int J Sports Phys Ther 2015;10:456-63. 31. Zuckerman SL, Chotai S, Devin CJ, Parker SL, Stonko DP, Wick JB, et al. Surgical resection of intradural extramedullary spinal tumors: patient reported outcomes and minimum clinically important difference. Spine 2016;41:1925-32. http://dx.doi.org/10.1097/BRS.0000000000001653
www.elsevier.com/locate/ymse
ORIGINAL ARTICLE
A preliminary exploration of plain-film radiography in scapular dyskinesis evaluation Kang Chen, MD1, Simin Deng, MD1, Yanhong Ma, MD*, Yelin Yao, MD, Juan Chen, MD, Yuqian Zhang, MD Department of Rehabiliation Medicine, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, Shanghai, China Background: Evaluation of scapular dyskinesis is of clinical interest because it is believed to be associated with pathologies of the shoulder. This study investigated the feasibility of plain-film radiography in evaluating scapular dyskinesis. Methods: Subjects with unilateral disorders of the shoulder (n = 186) who underwent plain-film radiography of bilateral scapulae were divided into 4 categories of scapular dyskinesis patterns according to the Kibler classification and analyzed. Coracoid upward shift distance (CUSD), length of the scapular spine line (LSS), and scapular upward rotation angle (SURA) were measured on the radiographs. Intrarater and inter-rater reliability were tested, and the characteristics of these parameters in each type were analyzed. The differences (d) between bilateral scapulae (d-CUSD, d-LSS, and d-SURA) among the 4 categories were compared. Results: Intrarater and inter-rater reliability were excellent for all parameters. Significant differences between the scapulae were observed in CUSD in type I and in LSS in type II categories. No significant difference in any of the parameters was found in type III. Compared with the other categories, d-CUSD in type I and d-LSS in type II were significantly larger. The cutoff values of d-CUSD and d-LSS were 1.1 mm and 1.2 mm, respectively. No significant difference in d-SURA was found among the 4 categories. Conclusions: The measurement of CUSD, LSS, and SURA on plain-film radiography had excellent reliability. d-CUSD and d-LSS were characteristic parameters of type I and type II, respectively; however, type III had no distinguishing characteristics among the parameters. Level of evidence: Level III; Diagnostic Study © 2018 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. Keywords: Scapular dyskinesis; plain-film radiography; coracoid upward shift distance; length of the scapular spine line; scapular upward rotation angle; Kibler classification
Abnormalities in the positions of the scapulae at rest and during movements have been collectively termed scapular dyskinesis,15,20,27 which is associated with many shoulder disorders.8,11,17-19,25 Management of scapular dyskinesis is one The Shanghai Jiao Tong University Affiliated Sixth People’s Hospital Ethics Committee approved the study: Approval No. 2015-43-(1). 1 These authors contributed equally to this work. *Reprint requests: Yanhong Ma, MD, Department of Rehabilitation Medicine, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, 600 Yishan Rd, Shanghai, 200233, China. E-mail address: [email protected] (Y. Ma).
of the strategies of physiotherapy in disorders of the shoulder; therefore, appropriate evaluation is important in the management of scapular dyskinesis. Kibler et al20 introduced a classification system involving 4 categories for scapular dyskinesis according to the abnormal movements of the scapula in 3 planes: Type I: abnormal scapular movements in the sagittal plane with anterior and posterior tilts Type II: abnormal scapular movements in the transverse plane with internal and external rotations
1058-2746/$ - see front matter © 2018 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved. https://doi.org/10.1016/j.jse.2017.12.032
ARTICLE IN PRESS 2
K. Chen et al. Type III: elevation of the scapula and abnormal movements in the coronal plane with upward and downward rotations Type IV: symmetrical bilateral scapula
This classification system is widely used in clinical settings15,16,20-22 but is based on visual observations.20 In many cases, the soft tissues overlying the scapulae, such as adipose tissue and muscles, may make the identification difficult,30 and this method cannot quantify the severity of scapular dyskinesis. Laboratory instruments, such as inclinometers and electromagnetic or optoelectronic tracking systems, can quantify scapular kinematics in the 3 dimensions.3,8,10,24,28 These methods allow for accurate measurements; however, the technical difficulties and requirements of the equipment render them unavailable in the clinical settings. Three-dimensional (3D) wing computed tomography (CT) and magnetic resonance imaging (MRI) have been explored in evaluating scapular dyskinesis.21-23 However, CT and MRI are both performed in the decubitus position, which relieves the scapula of gravity and the resulting movements. Whether scapular kinematics in decubitus are the same as those in the erect position is unclear. Endo et al 6,7 introduced 3 parameters on plain-film radiography to represent 3 kinds of scapular rotational tilts. The length of the scapular spine line (LSS) represents internal and external rotations, coracoid upward shift distance (CUSD) represents anterior and posterior tilts, and the scapular upward rotation angle (SURA) represents upward and downward rotations. Plain film is advantageous because of lower radiation, fewer technical requirements, and inexpensive and easily available equipment. We hypothesized that LSS, CUSD, and SURA can be used to evaluate scapular dyskinesis. Therefore, the purpose of this study was to investigate the characteristics of radiographic parameters in the 4 patterns of scapular dyskinesis and the feasibility of plain-film radiography in evaluating scapular dyskinesis.
Materials and methods Subjects and demographics This was a prospective, diagnostic case series involving repeated measurements. All subjects received information regarding the objectives of the study and provided their written consent for participation. Subjects who had unilateral shoulder disorders were recruited from the Department of Rehabilitation Medicine by 1 investigator (K.C.). Exclusion criteria were (1) a history of fracture or deformity of the scapula, clavicle, humerus, or ribs; (2) scoliosis or kyphosis with visible rib hump in neutral erect posture; (3) unequal lengths of lower extremities; (4) stature >1.8 m or body mass index >26 kg/m2;12 (5) subjects whose scapular pattern could not be determined by 3
Table I
Demographic data of subjects No. or mean ± SD (range)
Variable
(N = 186) Sex Women Men Age, yr Mass, kg Height, m BMI, kg/m2 Affected side Left Right Dominant side Left Right Dominant side Affected Unaffected Duration, weeks
115 71 45.74 ± 10.20 (22-68) 59.60 ± 6.94 (46-80) 1.64 ± 0.07 (1.50-1.80) 22.14 ± 1.93 (17.47-25.96) 73 113 15 171 110 76 16.82 ± 13.22 (1-48)
SD, standard deviation; BMI, body mass index.
Table II
Main disorders in the affected shoulders
Main disorder of affected side
No. (N = 186)
Subacromial impingement syndrome Partial-thickness rotator cuff tear Adhesive capsulitis of shoulder Long head of biceps tendon tendinitis Shoulder osteoarthritis Rotator cuff tendonitis Superior labrum anterior posterior tear
59 56 23 15 13 10 10
raters; and (6) obvious variations or bony spurs of the coracoid and acromion as seen on x-ray images. We enrolled 205 subjects who underwent radiographic examination and clinical classification of scapular dyskinesis. After the radiographic examination, 19 individuals were excluded for the following reasons: (1) obvious acromion spurs in 8 and morphologically variable coracoid processes in 2; (2) mild thoracic scoliosis in 3; and (3) the quality of imaging was inadequate in 6. Finally, 186 individuals were included, and their demographic data are summarized in Tables I and II.
Clinical evaluation for scapular dyskinesis Scapular patterns were identified based on visual palpation because it has the advantages of identification of bony landmarks and reliability.15 To expose both the scapulae, men bared to the waist and women wore halter tops. The raters stood behind the subject and documented the initial rating by visual inspection. Then they placed their hands on the subject’s scapulae for further identification
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation
3 posterior end of the acromion of the acromioclavicular joint.6,7 LSS was defined as the length of the scapular spine line (Fig. 2, A). CUSD was defined as the distance between the scapular spine line and the upper border of the coracoid process (Fig. 2, A). If the upper border of the coracoid process was above the scapular spine line, the value was denoted with a positive (+) sign and in other cases, a minus (–) sign was attached. The angle between the scapular spine line and the horizontal was defined as SURA (Fig. 2, B). If the LSS was above the horizontal, the value was denoted with a positive (+) sign and in other cases, it was denoted with a minus (–) sign.6,7
Reliability Figure 1 Schematic diagram of the visual-based palpation process in evaluating scapular dyskinesis.
(Fig. 1).15,16 The scapular patterns were classified according to the previously described Kibler classification.15,16,20,27 Three raters with previous experience with assessments of scapular dyskinesis4 assessed the subjects together and were allowed discussions to derive a unanimous decision. Before the study began, the raters were trained for 2 months using visual and verbal presentations, and they practiced the assessments in 10 subjects in a preliminary study. They had the same criteria for evaluation and were allowed discussions; therefore, they were able to provide agreements on all assessments. If a scapular pattern could not be determined because of a disagreement among the raters, the subject was excluded.
A priori power analysis showed that 22 was the minimum required number of participants to establish a reliability coefficient of 0.80 with α = 0.05 and 1 – β = 0.80.29 A computer-generated random number table was used to randomly select 22 radiographs from the 186 plain films for assessing the reproducibility of measurements. These measurements were observed by 2 measurers, with an interval of 1 week between the first and second measurements. The radiographic measurers were blinded to the results of the clinical evaluation.
Data processing The final value of a parameter was the average of 3 measurements. d-CUSD, d-LSS, and d-SURA were the differences in the values between the unaffected and affected sides.
Radiographic technique
Statistical analysis
Immediately after the clinical evaluation, subjects underwent anteroposterior (AP) radiography of both scapulae. The subject stood before the imaging plate, facing the x-ray tube, in the same erect position as that in the clinical evaluation. The sternal angle was selected as the radiographic center. The distance between the film and the x-ray tube was set at 1.5 m. The tube was directed perpendicular to the frontal axis of the thorax. All radiographies were obtained by 1 experienced image technologist.
SPSS 19.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA) software were used for statistical analyses. The normality of the quantitative data was tested using the Kolmogorov-Smirnov test. Intraclass correlation coefficient (ICC) was used to test intrarater and inter-rater reliability, which was denoted as poor (0.00-0.20), fair (0.21-0.40), good (0.41-0.75), or excellent (>0.75) reliability.9 The 95% confidence intervals (CIs) of the ICC of each parameter of the unaffected side and the affected side were computed. The standard error of measurement (SEM) and the minimal detectable change (MDC) for 95% CIs were calculated for each parameter using the following formula:
Radiographic parameters measurements Radiographic plain film was saved in DICOM (Digital Imaging and Communications in Medicine; National Electrical Manufacturers Association, Rosslyn, VA, USA) format in a picture archiving and communication system. Digimizer 4.3.5 software (MedCalc Software bvba, Ostend, Belgium) was used to measure the distances and angles.13 During the measurement, the radiographic digital images were magnified to clearly visualize the inner and outer points (Fig. 2). The right scapula was measured first, followed by the left. Each measurement was repeated 3 times. The measurer first marked a referential line, named the scapula spine line, on the image. The inner point of this line was the intersection of the upper border of the scapular spine and the floor of the supraspinatus fossa, and the outer point was the
SEM = SDpooled × 1 − ICC , MDC = SEM × 2 × 1.96, where SDpooled was the pooled standard deviation.14 The side-to-side differences in the radiographic parameters were tested by the paired Student test. One-way analysis of variance was used to compare d-CUSD, d-LSS, and d-SURA between the 4 types of scapular dyskinesis, and the least significant difference test was performed as a post hoc test when needed. The parameters with significant differences in d-CUSD and d-LSS on analysis of variance were further analyzed using the receiver operating characteristic (ROC) curves. Statistical significance was defined as a P value of < .05.
ARTICLE IN PRESS 4
K. Chen et al.
Figure 2 Schematic diagram of measuring the radiographic parameters. Anteroposterior radiograph of a shoulder at rest is shown. The red dots represent the inner and outer points of the scapular spine line. (A) Length of the scapular spine line (LSS) was defined as the length of the scapular spine line. Coracoid upward shift distance (CUSD) was defined as the distance between the scapular spine line and the upper border of the coracoid process. (B) Scapular upward rotation angle (SURA) was defined as the angle between the scapular spine line and the horizontal.
Results
IV patterns (Fig. 3). Representative radiographs are shown in Fig. 4.
Clinical evaluation In the 186 subjects, the numbers of subjects with scapular dyskinesis type I, II, III, and IV were 28, 41, 71, and 46, respectively.
Reliability of the radiographic measurement The reliability, SEM, and MDC of parameters are summarized in Tables III and IV. Intrarater and inter-rater reliabilities for all parameters in the unaffected and affected sides were excellent.
Side-to-side comparisons of the radiographic parameters
Parameter CUSD, mm LSS, mm SURA, °
Significant differences in d-CUSD and d-LSS were observed among the 4 types (P < .001). The post hoc test showed that d-CUSD in type I was significantly larger compared with the other types (type II: P = .001; type III: P < .001; type IV: P < .001; Fig. 5, A). d-LSS in type II was significantly larger than that in the other types (P < .001) (Fig. 5, B). There was no obvious difference in d-SURA among the 4 types (P = .254; Fig. 5, C).
Predictive values of d-CUSD for type I and d-LSS for type II
CUSD in type I and LSS in type II indicated a significant difference between the unaffected and affected sides (CUSD: 2.8 ± 2.5 mm; LSS: 3.9 ± 4.0 mm). None of the parameters showed significant side-to-side differences in type III and type Table III
Comparison of d-CUSD, d-LSS, and d-SURA among the 4 types
In view of the significant correlations shown by d-CUSD and d-LSS with type I and type II patterns, respectively, ROC curves were drawn to evaluate the predictive ability of these
Intrarater reliability of radiographic parameters in the affected and unaffected sides Side Affected Unaffected Affected Unaffected Affected Unaffected
Measurement 1
Measurement 2
Mean ± SD
Mean ± SD
0.5 ± 3.9 1.0 ± 4.4 70.7 ± 5.6 69.4 ± 6.4 7.5 ± 6.2 9.4 ± 7.6
0.5 ± 3.9 1.0 ± 4.3 70.1 ± 5.2 69.1 ± 6.2 8.4 ± 7.0 9.7 ± 7.5
ICC (95% CI)
SEM
MDC95
0.999 (0.998-1) 1 (0.999-1) 0.982 (0.957-0.993) 0.992 (0.980-0.997) 0.962 (0.909-0.984) 0.997 (0.992-0.999)
0.1 0 0.7 0.6 1.3 0.4
0.4 0 2.0 1.6 3.5 1.1
ICC, intraclass correlation coefficient; CI, confidence interval; SEM, standard error of measurement; MDC95, minimal detectable change for 95% confidence interval; SD, standard deviation; CUSD, coracoid upward shift distance; LSS, length of the scapular spine line; SURA, scapular upward rotation angle.
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation Table IV
5
Inter-rater reliability of radiographic parameters in the affected and unaffected sides
Parameter CUSD, mm LSS, mm SURA, °
Side Affected Unaffected Affected Unaffected Affected Unaffected
Measurer 1
Measurer 2
Mean ± SD
Mean ± SD
0.5 ± 3.9 1.0 ± 4.4 70.7 ± 5.6 69.4 ± 6.4 7.5 ± 6.2 9.4 ± 7.6
0.4 ± 4.2 1.0 ± 4.4 71.1 ± 5.8 69.9 ± 6.2 8.2 ± 7.1 9.8 ± 7.4
ICC (95%CI)
SEM
MDC95
0.992 (0.981-0.997) 1 (0.999-1) 0.982 (0.956-0.992) 0.989 (0.974-0.996) 0.958 (0.899-0.983) 0.996 (0.991-0.998)
0.4 0 0.8 0.7 1.4 0.5
1 0 2.1 1.8 3.8 1.3
ICC, intraclass correlation coefficient; CI, confidence interval; SEM, standard error of the measurement; MDC95, minimal detectable change for 95% confidence interval; SD, standard deviation; CUSD, coracoid upward shift distance; LSS, length of the scapular spine line; SURA, scapular upward rotation angle.
2 parameters. The cutoff values of ROC curves and area under curve (AUC) are reported in Table V. The parameter with values smaller than the cutoff value was considered to be negative, and the parameter with values larger than the cutoff was considered positive and diagnostically significant.
Discussion With the deep understanding of the relationship between scapular kinematics and shoulder disorders, it is imperative to study the treatment modality of scapular dyskinesis.10,18,19 Although assessment of symptoms or function is often used to evaluate the effectiveness of the treatment modality used in scapular dyskinesis,2,26 few quantitative methods are suitable for clinical evaluation of the changes in scapular dyskinesis. This study shows that plain-film radiography can be used to evaluate scapular dyskinesis. Plain-film radiography is a simple and feasible investigation. Although it is 2D imaging, CUSD, LSS, and SURA can be used to evaluate 3D rotational tilts of the scapula.6,7 In the scapula-lateral view, the distance between the upper end of the coracoid process and the lateral point of the scapular spine line can be considered to be constant.1 Photoshop software (Adobe Systems Inc., San Jose, CA , USA) can be used to rotate the digital image to make the scapular line horizontal.6 CUSD is equal to the product of this constant and the sine value of the anterior/posterior tilt angle in the sagittal plane, and the corresponding change in CUSD would depend on the anterior/posterior tilt angle (Fig. 6, A).7 From a superioraxial view, the length of the scapular spine can be regarded as a constant, LSS is equal to the product of this constant and the cosine value of the internal/external rotation angle, and the corresponding change of the LSS would depend on the internal/external rotation angle (Fig. 6, B).7 The upward or downward rotation of the scapula is in the coronal plane; therefore, the corresponding change in SURA would depend on the angle of the rotation.6,7 As described by Kibler, the type I pattern would be associated with abnormal anterior tilting of the scapula, and the type II pattern would be associated with abnormal scapular
internal rotation.20,27 Some previous studies that observed the scapular pattern using 3D wing CT have proved that subjects with type I exhibit increased anterior tilt and subjects with type II exhibit increased internal rotation.21,22 In this study, CUSD in type I and LSS in type II patterns were significantly smaller in the affected side compared with the unaffected side. Further testing of MDC for the 95% CI indicated that CUSD and LSS could significantly assess anterior tilt and internal rotation, respectively. Similar to the study by Endo et al,6 excellent ICC was achieved for the intrarater and inter-rater reliabilities for all radiographic parameters bilaterally in this study. This may be attributed to the experience of the measurer and standardization of the procedures. Satisfactory reliability is essential for clinical application of such inferences. The AUC reflects the accuracy of a measurement. An AUC of <0.5 is considered random association, whereas an AUC between 0.7 and 0.8 or between 0.8 and 0.9 is considered adequate or excellent.31 Our results have demonstrated that d-CUSD for type I and d-LSS for type II were reasonably accurate (AUC values were 0.783 and 0.752, respectively). Therefore, d-CUSD and d-LSS can be used as the characteristic parameters in the diagnosis of type I and type II patterns. In addition, our study found that when d-CUSD was 1.1 mm, its sensitivity was 0.821 and specificity was 0.69 for identifying type I and nontype I patterns. When d-LSS was 1.2 mm, its sensitivity was 0.756 and specificity was 0.731 for identifying type II and nontype II patterns. The results indicated that d-CUSD ≥1.1 mm and d-LSS ≥1.2 mm have diagnostic significance. Although the Kibler classification system has been accepted as the standard in the evaluation of scapular dyskinesis,15,20,21 some controversies still exist, especially regarding type III. Some investigators believe that type III is associated with excessive upward rotation of the scapula,5,15 whereas other investigators believe that the abnormal movement in type III occurs in multiple planes, including elevation, anterior displacement, or upward rotation.20-22,27 In our study, SURA, which represented upward rotation of the scapula, showed no significant change in subjects with
ARTICLE IN PRESS 6
K. Chen et al.
Figure 3 Side-to-side comparisons of the radiographic parameters for each type of scapular dyskinesis. The range bars show the standard error of measurement. CUSD, coracoid upward shift distance; LSS, length of the scapular spine line; SURA, scapular upward rotation angle.
type III pattern. Similarly, CUSD and LSS showed no significance in type III pattern. We therefore believe that the type III pattern is not simply presented as upward rotation, anterior tilt, or internal rotation of the scapula. The movement
pattern in type III may involve motions other than rotation. It may be necessary to develop other parameters or methods to describe the movements of the scapula in type III pattern or define subtypes in terms of the various abnormal planes
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation
7
Figure 4 Representative radiographs of each type of scapular dyskinesis. (A) Type I: Coracoid upward shift distance (CUSD) in the affected and unaffected sides is –6.1 mm and 2.9 mm, respectively. (B) Type II: Length of the scapular spine line (LSS) in the affected and unaffected sides is 54.7 and 63.6 mm, respectively. (C) Type III: Scapular upward rotation angle (SURA) in the affected and unaffected sides is 10.8 and 4.8°, respectively. (D) Type IV: CUSD and LSS in the affected and unaffected sides are –0.4 mm and –0.8 mm, and 61.6 mm and 61.1 mm, respectively. (E) Type IV: SURA in the affected and unaffected sides is 9.9° and 9.3°, respectively. The black ellipse indicates the affected side.
Figure 5 Comparison of difference (d) for coracoid upward shift distance (d-CUSD), length of the scapular spine line (d-LSS), and scapular upward rotation angle (d-SURA) among the 4 types of scapular pattern are shown. The range bars show the standard error of measurement.
of movements. According to the current findings, a scapular pattern that does not meet the diagnostic criteria of type I or type II may be classified as type III. Scapular dyskinesis can occur when stationary and in motion, only in motion, or only in a static state.4,20 The movements of the glenohumeral joint occur in various directions and are often complex2,19; therefore, the evaluation of scapular
abnormalities during motion is difficult. The current study only explored the evaluation for the scapular dyskinesis in a static position. Compared with dynamic abnormalities, static abnormalities mainly involve the tension and length of soft tissue, and dynamic abnormalities may also involve the abnormalities of the strength of the related muscle.4 One of the limitations of this method is that it is only suitable for evaluating scapular
ARTICLE IN PRESS 8
K. Chen et al. Table V Diagnostic accuracy of difference in coracoid upward shift distance and difference in length of the scapular spine line for specific types of scapular dyskinesis Parameter
Group vs. group
Cutoff value by ROC curve, mm
Sensitivity
Specificity
AUC (95% CI)
P
d-CUSD d-LSS
Type I vs. type II, III, IV Type II vs. type I, III, IV
1.1 1.2
0.821 0.756
0.69 0.731
0.783 (0.717-0.84) 0.752 (0.684-0.813)
<.001 <.001
ROC, receiver operating characteristic; AUC, area under the curve; CI, confidence interval; d-CUSD, difference in coracoid upward shift distance; d-LSS, difference in length of the scapular spine line.
Figure 6 Theoretical analysis of the radiographic parameters. (A) Schema of the coracoid upward shift distance (CUSD) is shown in the scapula-lateral view. The distance between the upper end of the coracoid process and the lateral point of the scapular spine line is labeled l. The line l and the horizontal plane formed the α angle (anterior/posterior angle). Because l was constant, sin α becomes the only variable that determines CUSD. (B) Schema of the length of the scapular spine line (LSS) in the superior-axial view. The length of the scapular spine was labeled n. The angle β (internal/external angle) was formed between the line n and the coronal plane. Because n could be regarded as a constant, cosine β becomes the only variable that determines LSS.
dyskinesis in static states, and patients with abnormalities during movements were not included. There were other limitations in the present study. First, the results of this study were based on a relative small sample of each pattern of scapular dyskinesis. A larger sample size is needed to confirm the universality of the findings in future studies. Second, this study investigated a side-to-side comparison, and the results cannot be applied to evaluate subjects who have bilateral shoulder disorders. Third, the stature and body mass index of subjects were limited because of the limited size of the imaging plate. Fourth, because a clear definition of the line of reference was essential for the measurements, subjects with obvious variations or bony spurs of the coracoid and acromion were excluded; therefore, populations suitable for present radiologic examination are limited.
Conclusion Type I and type II scapular dyskinesis can be evaluated with d-CUSD and d-LSS on plain-film radiography with excellent reliability. However, further explorations are warranted to understand the characteristics of type III.
Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References 1. Alobaidy MA, Soames RW. Evaluation of the coracoid and coracoacromial arch geometry on Thiel-embalmed cadavers using the three-dimensional MicroScribe digitizer. J Shoulder Elbow Surg 2016;25:136-41. http://dx.doi.org/10.1016/j.jse.2015.08.036 2. Carbone S, Postacchini R, Gumina S. Scapular dyskinesis and SICK syndrome in patients with a chronic type III acromioclavicular dislocation. Results of rehabilitation. Knee Surg Sports Traumatol Arthrosc 2015;23:1473-80. http://dx.doi.org/10.1007/s00167-014-2844-5 3. Chu Y, Akins J, Lovalekar M, Tashman S, Lephart S, Sell T. Validation of a video-based motion analysis technique in 3-D dynamic scapular kinematic measurements. J Biomech 2012;45:2462-6. http://dx.doi.org/ 10.1016/j.jbiomech.2012.06.025 4. Deng S, Chen K, Ma Y, Chen J, Huang M. The influence of test positions on clinical assessment for scapular dyskinesis. PM R 2017;9:761-6. http://dx.doi.org/10.1016/j.pmrj.2016.11.011 5. Ellenbecker TS, Kibler WB, Bailie DS, Caplinger R, Davies GJ, Riemann BL. Reliability of scapular classification in examination of professional
ARTICLE IN PRESS Plain-film radiography in scapular dyskinesis evaluation
6.
7.
8.
9. 10.
11.
12. 13.
14. 15.
16.
17. 18.
19.
baseball players. Clin Orthop Relat Res 2012;470:1540-4. http:// dx.doi.org/10.1007/s11999-011-2216-0 Endo K, Ikata T, Katoh S, Takeda Y. Radiographic assessment of scapular rotational tilt in chronic shoulder impingement syndrome. J Orthop Sci 2001;6:3-10. Endo K, Yukata K, Yasui N. Influence of age on scapulo-thoracic orientation. Clin Biomech (Bristol, Avon) 2004;19:1009-13. http:// dx.doi.org/10.1016/j.clinbiomech.2004.07.011 Fayad F, Roby-Brami A, Yazbeck C, Hanneton S, Lefevre-Colau MM, Gautheron V, et al. Three-dimensional scapular kinematics and scapulohumeral rhythm in patients with glenohumeral osteoarthritis or frozen shoulder. J Biomech 2008;41:326-32. http://dx.doi.org/10.1016/ j.jbiomech.2007.09.004 Fleiss J. The design and analysis of clinical experiments. New York: John Wiley & Sons; 1986. Forthomme B, Crielaard JM, Croisier JL. Scapular positioning in athlete’s shoulder : particularities, clinical measurements and implications. Sports Med 2008;38:369-86. Groh GI, Mighell MA, Basamania CJ, Kibler WB. All things clavicle: from acromioclavicular to sternoclavicular and all points in between. Instr Course Lect 2016;65:181-96. Gupta M, Dashottar A, Borstad JD. Scapula kinematics differ by body mass index. J Appl Biomech 2013;29:6. Habibi E, Soury S, Zadeh AH. Precise evaluation of anthropometric 2D software processing of hand in comparison with direct method. J Med Signals Sens 2013;3:256-61. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 2000;30:1-15. Huang TS, Huang HY, Wang TG, Tsai YS, Lin JJ. Comprehensive classification test of scapular dyskinesis: a reliability study. Man Ther 2015;20:427-32. http://dx.doi.org/10.1016/j.math.2014.10.017 Huang TS, Ou HL, Huang CY, Lin JJ. Specific kinematics and associated muscle activation in individuals with scapular dyskinesis. J Shoulder Elbow Surg 2015;24:1227-34. http://dx.doi.org/10.1016/ j.jse.2014.12.022 Kibler WB. The scapula in rotator cuff disease. Med Sport Sci 2012;57:27-40. http://dx.doi.org/10.1159/000328877 Kibler WB, Sciascia A. The role of the scapula in preventing and treating shoulder instability. Knee Surg Sports Traumatol Arthrosc 2016;24:390-7. http://dx.doi.org/10.1007/s00167-015-3736-z Kibler WB, Sciascia A, Wilkes T. Scapular dyskinesis and its relation to shoulder injury. J Am Acad Orthop Surg 2012;20:364-72. http:// dx.doi.org/10.5435/JAAOS-20-06-364
9 20. Kibler WB, Uhl TL, Maddux JW, Brooks PV, Zeller B, McMullen J. Qualitative clinical evaluation of scapular dysfunction: a reliability study. J Shoulder Elbow Surg 2002;11:550-6. http://dx.doi.org/10.1067/ mse.2002.126766 21. Park JY, Hwang JT, Kim KM, Makkar D, Moon SG, Han KJ. How to assess scapular dyskinesis precisely: 3-dimensional wing computer tomography–a new diagnostic modality. J Shoulder Elbow Surg 2013;22:1084-91. http://dx.doi.org/10.1016/j.jse.2012.10.046 22. Park JY, Hwang JT, Oh KS, Kim SJ, Kim NR, Cha MJ. Revisit to scapular dyskinesis: three-dimensional wing computed tomography in prone position. J Shoulder Elbow Surg 2014;23:821-8. http://dx.doi.org/ 10.1016/j.jse.2013.08.016 23. Sahara W, Sugamoto K, Murai M, Yoshikawa H. Three-dimensional clavicular and acromioclavicular rotations during arm abduction using vertically open MRI. J Orthop Res 2007;25:1243-9. http://dx.doi.org/ 10.1002/jor.20407 24. Shakeri H, Keshavarz R, Arab AM, Ghosheh FT, TalimKhani A. Scapular position and orientation during abduction, flexion and scapular plane elevation phase. Iran Rehabil J 2014;12:22-30. 25. Timmons MK, Thigpen CA, Seitz AL, Karduna AR, Arnold BL, Michener LA. Scapular kinematics and subacromial-impingement syndrome: a meta-analysis. J Sport Rehabil 2012;21:354-70. 26. Turgut E, Duzgun I, Baltaci G. Effects of scapular stabilization exercise training on scapular kinematics, disability, and pain in subacromial impingement: a randomized controlled trial. Arch Phys Med Rehabil 2017;98:1915-23, e1913. http://dx.doi.org/10.1016/j.apmr.2017.05.023 27. Uhl TL, Kibler WB, Gecewich B, Tripp BL. Evaluation of clinical assessment methods for scapular dyskinesis. Arthroscopy 2009;25:12408. http://dx.doi.org/10.1016/j.arthro.2009.06.007 28. van den Noort JC, Wiertsema SH, Hekman KM, Schonhuth CP, Dekker J, Harlaar J. Reliability and precision of 3D wireless measurement of scapular kinematics. Med Biol Eng Comput 2014;52:921-31. http:// dx.doi.org/10.1007/s11517-014-1186-2 29. Walter SD, Eliasziw M, Donner A. Sample size and optimal designs for reliability studies. Stat Med 1998;17:101-10. 30. Wassinger CA, Williams DA, Milosavljevic S, Hegedus EJ. Clinical reliability and diagnostic accuracy of visual scapulohumeral movement evaluation in detecting patients with shoulder impairment. Int J Sports Phys Ther 2015;10:456-63. 31. Zuckerman SL, Chotai S, Devin CJ, Parker SL, Stonko DP, Wick JB, et al. Surgical resection of intradural extramedullary spinal tumors: patient reported outcomes and minimum clinically important difference. Spine 2016;41:1925-32. http://dx.doi.org/10.1097/BRS.0000000000001653